The copyright © of this thesis belongs to its rightful author and/or other copyright owner. Copies can be accessed and downloaded for non-commercial or learning purposes without any charge and permission. The thesis cannot be reproduced or quoted as a whole without the permission from its rightful owner. No alteration or changes in format is allowed without permission from its rightful owner.
POLITICAL INFLUENCES AND EARNINGS QUALITY:
EVIDENCE FROM PAKISTAN
MUHAMMAD SADIQ
DOCTOR OF PHILOSOPHY UNIVERSITI UT ARA MALAYSIA
2018
POLITICAL INFLUENCES AND EARNINGS QUALITY:
EVIDENCE FROM PAKISTAN
By
MUHAMMAD SADIQ
Thesis Submitted to
Tunku Puteri Intan Safinaz School of Accountancy Universiti Utara Malaysia
in Fulfillment of the Requirement for the Degree of Doctor of Philosophy
TUNKU PUTERI INTAN SAFINAZ SCHOOL OF ACCOUNTANCY
COLLEGE OF BUSINESS Universiti Utara Malaysia
PERAKUAN KERJA TESIS / DISERTASI (Certification of thesis I dissertation)
Kami, yang bertandatangan, memperakukan bahawa (We, the undersigned, certify that)
calon untuk ljazah
(candidate for the degree oQ
MUHAMMAD SADIQ DOCTOR OF PHILOSOPHY
telah mengemukakan tesis / disertasi yang bertajuk:
(has presented his/her thesis I dissertation of the following title):
POLITICAL INFLUENCES AND EARNINGS QUALITY: EVIDENCE FROM PAKISTAN
seperti yang tercatat di muka surat tajuk dan kulit tesis / disertasi.
(as it appears on the title page and front cover of the thesis I dissertation).
Bahawa tesis/disertasi tersebut boleh diterima dari segi bentuk serta kandungan dan meliputi bidang ilmu dengan memuaskan, sebagaimana yang ditunjukkan oleh calon dalam ujian lisan yang diadakan pada:
17 Oktober 2017.
(That the said thesis/dissertation is acceptable in form and content and displays a satisfactory knowledge of the field of study as demonstrated by the candidate through an oral examination held on:
17 October 2017.
Pengerusi Viva (Chairman for Viva)
Pemeriksa Luar (External Examiner)
Pemeriksa Dalam (Internal Examiner)
Tarikh: 17 October 2017 (Date)
Assoc. Prof. Dr. Chek Derashid Tanc i a t a n g a ~ ~ ~ : t _ _ _ _ _ _ _ __ __ _ _ _ __ _ (Signature)
Tandatangan
7 4 ~ ,.
(Signature)
- - -
Prof. Dr. Norman Mohd Sale- - - -- -
h (UKM- --
)- - - -
Tandatangan _D_r_. M_o_h_d_. 'A_t_ef_M_d_Y_u_so_f _ __ _ __ __ (Signature)
Nama Pelajar {Name of Student}
Tajuk Tesis / Disertasi
(Tftle of the Thesis I Dissertation)
Program Pengajian (Programme of Study)
Nama Penyelia/Penyelia-penyelia (Name of Supervisor/Supervisors)
Nama Penyelia/Penyelia-penyelia (Name of Supervisor/SupeNisors)
Muhammad Sadiq
POLITICAL INFLUENCES AND EARNINGS QUALITY: EVIDENCE FROM PAKISTAN
Doctor of Philosophy
Assoc. Prof. Dr. Zaleha Othman
Dr. Rohami Shafie
Tanda/; n r
PERMISSION TO USE
In presenting this thesis in fulfillment of the requirements for a postgraduate degree from Universiti Utara Malaysia, I agree that the Universiti Library may make it freely available for inspection. I further agree that permission for the copying of this thesis in any manner, in whole or in part, for scholarly purpose may be granted by my supervisor(s) or, in their absence, by the Dean of Tunku Puteri Intan Safinaz School of Accountancy where I did my thesis. It is understood that any copying, publication, or use of this thesis or parts thereof for financial gain shall not be allowed without my written permission. It is also understood that due recognition shall be given to me and to Universiti Utara Malaysia (UUM) for any scholarly use which may be made of any material from my thesis.
Requests for permission to copy or to make other use of materials in this thesis, in whole or in part should be addressed to:
Dean of Tunku Puteri In tan Safinaz School of Accountancy Universiti Utara Malaysia
060 l 0 UUM Sintok Kedah Darul Aman
iv
ABSTRACT
The general objective of this study was to examine the relationship between political influences and earnings quality. Specifically, this study examined the relationship between the ruling party politically connected firms and poor earnings quality. In addition, the study also compared the opposition party politically connected finns, when politically influenced firms substitute real earnings management for accrual earnings management in Pakistan. This study used the data of 129 firms listed on the Pakistan Stock Exchange over the period 2009-2013. The Panel Corrected Standard Error (PCSE) technique was employed to solve the heteroskedasticity issue. The results showed that politically influenced firms are reporting poor earnings quality. Contrary to previous studies, politically connected firms through opposition party(s) report poor earnings quality compared to ruling party(s) politically connected firms, when earnings quality is measured by the real earnings management attribute of earnings quality.
However, there is no difference between the ruling party and the opposition party political connections on earnings quality, when earnings quality is measured by the accrual earnings management attributes of earnings quality. The study also found that politically influenced firms substitute real earnings management for accrual earnings management. Further, this study revealed that firms with stronger political connections are more involved in accrual earnings management activities and firms with weaker political connections are more inclined to use the real earnings management strategy.
Accordingly, regulators must keep in mind political factors during regulatory reforms.
This study provides detailed investigations and deep insight on the relationship between political influences and earnings quality. This study contributes to the field of earnings management where it integrates the agency theory with the political economy theory.
Apart from that, this study contributes to practice where it provides deep insight to policy-makers who are interested in improving corporate governance in Pakistan.
Keywords: political influences, earnings quality, earnings management, Pakistan.
ABSTRA.K
Objektif am kajian ini adalah meneliti hubungan antara pengaruh politik dan kualiti pendapatan. Secara khususnya, kajian i.Jri meneliti hubungan antara finna yang mempunyai kaitan dengan parti politik pemerintah dan kualiti pendapatan yang lemah.
Di samping itu, kajian ini juga membuat perbandingan antara firma yang mempunyai kaitan dengan politik dengan parti lawan di Pakistan. Jni apabila finna menggantikan pengurusan pendapatan pilihan dan apabila firma mengunakan pengurusan pendapatan akruan. Kajian ini menggunakan data daripada 129 buah firma yang disenaraikan dalam Bursa Saham Pakistan dari tahun 2009 hingga2013. Teknik Panel Corrected Standard Error (PCSE) digunakan untuk menyelesaikan masalah heteroscedasticity. Kajian mendapati firma yang mempunyai pengaruh politik melaporkan kualiti pendapatan yang lemah. Hal ini berbeza dengan kajian terdahulu iaitu firma yang mempunyai pengaruh politik daripada pihak la wan melaporkan kualiti pendapatan yang lemah apabila kualiti pendapatan diukur menggunakan pengurusan pendapatan sebenar sebagai atribut kualiti pendapatan. Waiau bagaimanapun, kajian juga mendapati tiada kesan perhubungan antara parti pemerintah dan parti lawan apabila kualiti pendapatan diukur menggunakan pengurusan pendapatan akruan sebagai atribut kualiti pendapatan. Kajian juga mendapati firma yang mempunyai pengaruh politik, melaporkan pengurusan pendapatan sebenar berbanding dengan pengurusan pendapatan akruan. Justeru6, kajian juga mendedah.kan bahawa firma yang mempunyai pengaruh politik yang kuat, melibatkan diri dengan aktiviti pengurusan pendapatan akruan dan firma yang mempunyai pengaruh politik yang lemah lebih cenderung kepada pengunaan strategi pengurusan pendapatan sebenar. Oleh itu, pihak berkuasa perlu mengambil perhatian akan hubungan politik apabila membuat reformasi terhadap penguatkuasaan. Secara keseluruhannya, kajian ini memberi penelitian yang terperinci dan mendalam akan perhubungan antara foma politik dan pengurusan pendapatan. Kajian ini menyumbang kepada pengurusan pendapatan apabila menghubungkaitkan teori agensi dengan teori ekonomi politik. Selain itu, kajian ini juga menyumbang kepada praktis apabila memberi maklumat yang mendalam kepada pembuat polisi yang berminat dalam menambah baik tataurus tadbir di Pakistan.
Kata kunci: pengaruh politik, kualiti pendapatan, pengurusan pendapatan, Pakistan
vi
ACKNOWLEDGEMENT
All praises are due to Allah, the almighty, for giving me the strength to complete my Doctor of Philosophy. I am greatly indebted to my supervisors Associate Professor Dr.
Zaleha Othman and Dr. Rohami Shafie for their encouragement and support throughout all stages of my PhD journey.
My sincere gratitude goes to the management of University Utara Malaysia for the Ph.D. scholarship grant; the scholarship grant indeed reduced my financial burden and aided the timely completion ofmy study.
I acknowledge the spiritual, moral, and financial support received from my parents:
Abdul Hameed Khan and Hayat Bibi. I would also like to thank my wife, Sabira Ali, my children Ahmed and Mustafa for their patience and prayers during my stay in Malaysia. In addition, I would like to thank my siblings, Shazia Adnan, Abdul Khaliq, Abdul Malik, Abdul Razik, Abdul Wahab, and Hafsa Hameed, for all their support and prayers during my absence. In addition, I would like to express my appreciation to my friends most especially Syed Ehsanullah, Ammar Mahmood Khan, Syed Hasan Ali, Ahsan Rafe, Fareed Butt, Atif Aziz, for the words of encouragement and advice.
Finally, my great thanks to Dr. Salau Abdul Malik, who was instrumental in developing my understanding of data analysis. I also owe a great deal of tJ1anks to my friend, Muhammad Haseeb Bhatti, for his unending advice and help in the entire process.
TABLE OF CONTENTS
TITLE PAGE I
CERTIFICATION OF THESIS II
PERMISSION TO USE IV
ABSTRACT V
ABSTRAK VI
ACKNOWLEDGEMENT VII
TABLE OF CONTENTS VIII
LIST OF TABLES XII
LIST OF ABBREVIATIONS XIV
LIST OF FIGURES XV
LIST OF APPENDICES XVI
CHAPTER ONE INTRODUCTION 1
1.1 Background 1
1.2 Problem Statement 10
I .3 Research Questions 13
1 .4 Research Objectives 14
1.5 Significance 15
l .6 Scope of the Study 20
1.7 Organization of the Study 20
CHAPTER TWO LITERATURE REVIEW 22
2.1 Earnings Quality Definitions 22
2.2 Measurements of Earnings Quality 24
2.2.1 Earnings Management 25
2.2. l. l Real Earnings Management 28
2.2.1.2 Accrual-Based Earnings Management 33
2.2.1.3 Accruals Quality 34
2.3 The difference between Accruals Earnings Management and Real Earnings
Management 36
2.4 Earnings Management Motivations 38
2.4.1 Earnings Benchmarks 38
2.4.2 Timing Equity Offerings 38
2.4.3 Executive Compensation 40
2.4.4 Debt Contracts 40
2.4.5 Political Cost 42
2.5 Earnings aggressiveness 43
2.6 Comparison of Measurements 44
2.7 Determinants of Earnings Quality 46
2.8 Political influences 52
2.8.1 The Politics in Pakistan's Context 56
2.8.2 Consequences of Political Influences 58
2.8.3 Impact of Political influences on earnings quality 67
viii
2.9 Earnings Management Motivations in Politically Influenced Firms 2.10 Control Variables
2.10.1 Firm Size 2.10.2 Audit Quality 2.10.3 Leverage
2.10.4 Return on Assets (ROA) 2.10.5 Loss
2.10.6 Growth
2.11 Underpinning Theory 2.11.1 Agency Theory 2.12 Supporting Theory
2.12.1 Classical Political Economy Theory 2.13 Literature Gap
2.14 Summary
76 77 78 78 79 80 80 81 81 81 85 85
87
89
CHAPTER THREE METHODOLOGY 91
3. 0 Introduction 91
3.1 Theoretical Framework 91
3.2 Hypothesis Development 96
3.2. l Political Influences and Accruals Earnings Management 96 3.2.2 Political Influences and Real Earnings Management 98 3.2.3 Political Influences and Earnings Aggressiveness 99 3 .2.4 Ruling/Opposition Party(s) Political Connections and Accruals Earnings
Management I 00
3.2.5 Ruling/Opposition Party(s) Political Connections and Real Earnings
Management I 00
3.2.6 Ruling/Opposition Party(s) Political Connections and Earnings Aggressiveness 101 3.2.7 Substitution of Real Earnings Management Strategy for Accruals
Management
3 .3 Measurement of Variables 3.3.1 Accruals Quality
3.3.2 Real earnings management 3.3.3 Earnings aggressiveness 3 .3 .4 Control Variables 3.3 .5 Political Influences
3.3.5. l Types of Political Influences 3 .4 Data collection and sample specification
3.4.1 Sources of data 3.4.2 Data sample
3.5 Diagnostic Test of Panel Data Analysis 3 .5 .1 Heteroskedasticity
3.5.2 Autocorrelation 3.5.3 Multicollinearity
Earnings 102 103 103 105 109 110 113 116 l I 9 119 120 122 122 123 123
3.7 Conclusion 132
CHAPTER FOUR DATA ANALYSIS 133
4.1 Introduction 133
4.2 Industry Classification 133
4.3 Descriptive Statistics 134
4.3.1 Dependent Variables 134
4.3.2 Independent Variables 135
4.3.3 Control Variables 136
4.4 Analysis of Pearson Correlation Matrix 137
4.5 Diagnostic Test Results 141
4.5.1 Heteroskedasticity Results 141
4.5.2 Autocorrelation Results 142
4.5.3 Multicollinearity Results 143
4.6 Measurements of Relationships 144
4.7 Panel Regression Results for Earnings Quality Attributes 145 4. 7 .1 Relationship between Political Influences ( combined proxy and multiple proxies) and Accruals Earnings Management (Accruals Quality) 146 4.7.2 Relationship between Political Influences (combined proxy and multiple
proxies) and Real Earnings Management 151
4. 7.3 Relationship between Political Influences ( combined aspect and multiple aspects) and earnings opacity (earnings aggressiveness) 156 4.7.4 Relationship between Political Connections (through ruling/opposition political party) and accruals earnings management I 60 4.7.5 Relationship between Political Connections (through ruling/opposition
political party) and real earnings management 162
4.7.6 Relationship between Political Connections (through ruling/opposition
political party) and earnings aggressiveness 164
4.7.7 Relationship between the interaction effect of political influences and accruals earnings management on real earnings management 167 4.8 Alternative measures of Earnings Quality Attributes 169 4.8.1 Sensitivity Analysis for Accruals Earnings Management Model (Accruals Quality) and Political Influences (using single aspect) 170 4.8.2 Sensitivity Analysis for Accruals Earnings Management Model (Accruals Quality) and Political Influences (using multiple proxies) 172 4.8.3 Sensitivity Analysis for Real Earnings Management Model (using individual measure) and Political Influences (using single aspect) 174 4.8.4 Sensitivity Analysis for Real Earnings Management Model (using individual measures) and Political Influences (using multiple aspects) 176 4.8.5 Sensitivity Analysis for the interaction of political influences and accruals earnings management on real earnings management 178 4.8.6 Sensitivity Analysis using different aspects of political influences individually 180 4.8.7 Sensitivity analysis for the relationship between direct and indirect political
connections and earnings quality attributes 184
X
CHAPTER FIVE FINDINGS AND DISCUSSION 5.0 Introduction
5.1 Overview of the Study 5.2 Overview of Results 5.3 Discussion
5.3. l Political Influences and Earnings Quality CHAPTER SIX CONCLUSION
6.1 Contributions of the Study 6.1 . I Theoretical Contributions 6.1 .2 Practical Contributions 6.2 Limitations of the Study 6.3 Future Research
6.4 Conclusion REFRENCES
187
187 187 189 192 197
204 204 204 207 210 213 214
216
Table 2.1 Table 2.2 Table 3.1 Table 3.2 Table 4.1 Table 4.2 Table 4.3 Table 4.4 Table 4.5 Table 4.6 Table 4.7 Table 4.8 Table 4.9 Table 4.10 Table 4.11 Table4.12 Table 4.13 Table 4.14 Table 4.15 Table 4.16 Table 4.17 Table 4.18 Table 4.19 Table 4.20 Table 4.21
LIST OF TABLES
Overview of Earnings Quality Measurements. 46 The review of politically influenced fmns: Major results of 72 empirical studies
Variables Definition 111
Types of political influences used in previous studies 117
Industry Classification 133
Descriptive Statistics 136
Coefficient correlations between variables 140 VIF results of political influences (using multiple aspects) and 143 control variables
Regression results for Political Influences and Accruals Earnings 150 Management
Regression results for Political Influences and Real earnings 155 Management
Regression results for Political Influences and Earnings 159 Aggressiveness
Regression results for Ruling/Opposition Party Political 161 Connections and Accruals Earnings Management
Regression results for Ruling/Opposition Party Political 163 Connections and Real Earnings Management
Regression results for Ruling/Opposition Party Political 166 Connections and Earnings Aggressiveness
Regression results for the Interaction effect of political influences 168 and accruals earnings management on real earnings management
Alternative measures of accruals earnings management with 171 combined aspect of political influences
Alternative measures of accruals earnings management with 173 multiple aspects of political influences
Alternative measures of real earnings management with combined 175 aspect of political influences
Alternative measures of real earnings management with multiple 177 aspects of political influences
Alternative results for the interaction effect of political influences 179 and accruals earnings management on real earnings management
Regression results of multiple aspects of political .influences 181 (using individually) and accruals earnings management
Regression results of multiple aspects of political influences 182 (using individually) and real earnings management
Regression results of multiple aspects of political influences 183 (using individually) and earnings aggressiveness
Regression results of direct/indirect political connections, and real 185 and accruals earnings management
Regression results of direct/indirect political connections, and 186
xii
Table 5.1 Table 5.2 Table 5.3 Table 5.4 Table 5.5
. .
eanungs aggressiveness
Summary of Hypotheses Testing for Political Influences and 190 Accruals Earnings Management
Summary of Hypotheses Testing for Political Influences and Real 190 Earnings Management
Summary of Hypotheses Testing for Political Influences and 190 Earnings Aggressiveness
Summary of Hypotheses Testing for Ruling/Opposition Party 191 Political Connections and Earnings Quality models
Summary of Hypothesis Testing for Politically Influenced firms 191 and Substitution for Earnings Management Strategies
AEM BIG4
BUR
EA EQ FASB GAAP GOVT IFRS PC PCR PCSE PI REM SECP SOX
LIST OF ABBREVIATIONS
Accrnals Earnings Management
The 4 largest audit firms worldwide, Deloitte, .KMPG, PricewaterhouseCoopers, and Ernst & Young
Presence of Civil/Military Bureaucrat( s) on Board of Directors Earnings Aggressiveness
Earnings Quality
Financial Accounting Standards Board Generally Accepted Accounting Principles Government Owned Firms
International Financial Reporting Standards Political Connections
Political Connections through Ruling Party Panel-Corrected Standard Errors
Political Influences
Real Earnings Management
Security Exchange Commission of Pakistan Sarbanes-Oxley Act, the United States
xiv
Figure 2.1 Figure 3.1
LIST OF FIGURES
Determinants of earnings quality examined by previous studies.
Theoretical Framework
51 95
LIST OF APPENDICES
Appendix A List of firms used in the current study 243
xvi
I.I Background
CHAPTER ONE INTRODUCTION
The top management of an organization communicates firms' performance to its stakeholders through financial reports or annual reports, which allows the stakeholders to differentiate between good perfonning and poor performing firms. However, this can only be reliable guide when the firms are reporting their earnings in a credible or reliable way (Dechow and Schrand, 2004). The managers communicate about firm performance in financial reporting, and accounting standards allowed managers to use their judgments in reporting. This gives discretions to the managers to choose reporting technique that tmly reflects the economic transactions of business. However, the discretions may also give opportunities to managers to misuse it, and get involved in earnings management activities (Healy and Wahlen, 1999).
Schipper ( 1989) defines earnings management ( earnings manipulation) activities as the purposeful intervention in financial reporting, with the intent of getting some private gain. There are also empirical evidences that highlight some factors that motivate finns
to get involved in earnings management. For example, Dechow and Schrand (2004) highlight factors which encourage finns to get engaged in managing earnings; such as issuing shares at high prices, meeting analysts and investors' expectations and reducing cost of acquiring new capital etc. In addition, Healy and Wahlen (1999) have also discussed several methods adopted by the finns to managed earnings i.e. overstating
assets, revenues and profits, understating earnings per share, premature revenue recognition, aggressive capitalization, changing depreciation and amortization policies, undetreporting expenses and liabilities, fictitious revenue recognition and overstating cash flows. Previous studies show that earnings management activities decrease the earnings quality.
Earnings quality is contextual and it means different things to different people. Prior studies exemplify that high earnings quality contain small amount of accruals (Sadiq and Othman, 2017; Bhattacharya, Daouk, and Welker, 2003), give a useful measure for assessing firms values (Dechow and Schrand, 2004 ), earning is predictable and persistent (Penman and Zhang, 2002). On the contrary, earnings quality problem arises when firms get involved in earnings management activities (Al-dhamari and Ismail, 2015), and report less transparent earnings (Bhattacharya, Daouk, and Welker, 2003).
The significance of high earnings quality bas long been recognized since still there are many firms that are involved in earnings management activities, and reporting poor earnings quality. For example, Toshiba and Valeant Phannaceutical was reported to be involved in activities like overstating earnings, postponing reporting losses, moving current year costs to later years, and inflating sales (Fortune, 2015). Other infamous earnings management scandals include Penn West Petroleum Ltd., Tesco, Satyam, Tyco international and HealthSouth. Penn West Petroleum Ltd moved millions of dollars from revenue expenditure to capital expenditure, which artificially reduced 20 percent of their operating expenses (Reuters, 2017). Tesco was found involved in earnings
2
management practices by reporting inflated profits (Telegragh, 2017). Satyam was involved in overstating its reported earnings, assets and also paying salaries to non- existing staff for several years. WorldCom on the other hand used a simple method to capitalize more than $11 Billion of expenditures as assets rather than expenses. It was also involved in underreporting expenses and inflating revenues through fake accounting entries. A Switzerland based company named as Tyco international was found involved in inflating income by its CEO and CFO. In Australia, HealthSouth was caught by Security exchange commission which was found inflating $1.4Billion earnings to meet its shareholders expectations.
Pakistan is not excluded in respect of earnings quality; scandals have engrossed finns such as Tandlianwala Sugar Mills, Chenab Limited, Callmate Ltd, NICL and OGRA are all linked to earnings quality issues (Dawn, 2015; Tribune, 2014; SECP, 2010; Dawn, 2008). The earnings quality gets worse when finns are politically influenced (Sadiq and Othman, 2017; Al-d.hamari and Ismail, 2015). A firm is considered as politically influenced in presence of politician(s) or close relative(s) of a politician(s) in a board or senior management; a firm is significantly owned by the government; and in presence of civil/military bureaucrat(s) in a board.
Security Exchange Commission of Pakistan (SECP) reported that a politically influenced firm named Chenab Limited was involved in manipulating share prices or artificially increasing share prices (SECP, 2010). On the other hand, Dawn newspaper reports that National Accountability Bureau (NAB) of Pakistan has started investigating
against a politically connected firm named Tandlianwala Sugar Mills (Dawn, 2015).
NAB accused Tandlianwala Sugar Mills for involving in suspicious transactions.
Similarly, CallMate was found involved in fraudulent practices such as reporting fictitious revenue or manipulated revenue (Dawn, 2008).
1n addition, a politically influenced firm named National Insurance Company Limited (NICL) was involved in buying properties more than the market value by giving benefits to politically connected people (Dawn, 2015). Further, the former Chairman and Directors of Oil and Gas Regulatory (OGRA) were engaged in manipulating earnings and share prices of Sui Northern Gas Pipeline (SNGPL) and Sui Southern Gas Company (SSGC) (Tribune, 2014). The Express tribune also reported that former Prime Minister Yousuf Raza Gillani and minister Raja Pervaiz Ashraf were also summoned for illegally appointing the chairman of OGRA. All these facts show that there is a strong nexus between politics and businesses in Pakistan.
Realizing the seriousness of earnings quality issues, regulatory bodies from many developed and developing countries realized and revamped the laws related to accounting practices. For example, the implementation of Sarbanes-Oxley Act (SOX) in 2002 was one immediate response to earnings quality soon after Enron scam. Some authors opine that accrual based earnings management (AEM) was reduced substantially after the implementation of SOX (Braam, Nandy, Weitzel, and Lodh, 2015; Cohen, Dey, and Lys, 2008), indicating the success of the revamping of the rules and regulations related to earnings management.
4
Prior studies have determined factors that influence the quality of earnings such as ownership and capital structure (Ben-Nasr, Boubakri and Cosset, 2015; Soderstrom and Sun, 2007), accounting methods (Altamuro, Beatty, and Weber, 2005), corporate governance (Zhao, Chen, Zhang and Davis, 2012; Lin and Hwang, 2010; Cohen, Krishnamoorthy, and Wright, 2004), corporate social responsibility (Kim, Park, and Wier, 2012), audit quality (Lin and Hwang, 2010; Prawitt, Smith, and Wood, 2009).
Significantly, some authors opine that corporate political influences are a global phenomenon and it creates complex economic consequences (Bleibtreu and Konigsgruber, 2015).
Previous studies even documented that earnings quality gets worse when firms are politically influenced (Chi, Liao, and Chen, 2016; Liu, Li, Zeng, and An, 20 l 6; Al- dhamari and Ismail, 2015; Liu, Saidi and Bazaz, 2014; Narayanaswamy, 2013; Chaney, Faccio, and Parsley, 2011 ). Ironically, in recent years political influences have become a common phenomenon (Saeed, Belghitar, and Clark, 2015), which is evident in the study of Zang (20 I I), who disclosed that 54.5 percent of the firms in their sample had politically influenced senior management in 2007 compared to 31.5 percent in 1990.
Such trend is also noted in developing countries where political connections are valuable for firms operating in economies where institutions are weak and politicians have no fear of accountability.
Fisman (2001) shows how finns that have close connections with politically connected people get favors. He cites the example that before the news of Indonesian President Suharto's bad health was publicly released, firms linked with President Suharto's family had already known about it while other viewed it as spread of rumors. Roberts (1990) finds out that in 1983 finns that contributed in elections campaign of Senator Henry Scoop Jackson, unexpectedly find their stock returns dropping after his sudden death.
At the same time, firms connected to Senator Sam witnessed unexpected increase in stock returns who was the successor of Senator Henry.
Moreover, it is also noticed that in developing economies politically influenced finns get favored bank loans at a lower interest rate and higher debt ratio. This view is supported by a number of empirical evidences. For instance, Charumilind (2006) find that political influences are an important determinant in Thailand to get a long-term financial leverage. Similarly, K.hwaja and Mi.an (2005) reveal that politically influenced firms are determined by a lower interest rate and higher debt equity ratio. Some researchers argue that political connection negatively affects earnings quality (for example, Al-dhamari and Ismail, 2015; Chaney, Faccio, and Parsley, 2011 ), while other researchers opine that politically influenced firms have additional power beyond country specific, regulatory and firm-specific ownership characteristics (e.g. Faccio, Masulis, and McConnell, 2006).
There is also empirical evidence proving that political influences are more common in countries with high corruption and less common in countries where strict regulations are
6
practiced and there exists political conflicts of interest (Faccio, 2006). Pakistan for instance, labeled as one of the most corrupted countries in Asia, is associated with corruption, power abuse and bribery (Saeed, Belghitar, and Clark, 2015). As a matter of fact, the Index of Economic Freedom has named Pakistan as amongst the top corrupt countries of the world (Saeed, Belghitar, and Clark, 2015). As evidence, during the last 25 years three elected prime ministers have been removed and four elected parliaments have been dissolved on charges of corruption and political abuse of power. Since the independence of Pakistan, business and politics in Pakistan have a very close association (Rehman, 2006). The first Chief Minister of Sindh province named Yousuf Haroon was one of the leading industrialists, and people like Naseer Shiekh and Rafiq Saigol were the founding owners of Saigol Group, and Ahmad Dawood (owner of Dawood group) held main positions in the governments (Rehman, 2006). In the general elections held in 2002, 2008 and 20 I 3, the political parties that participated were mainly led by politicians representing some of the country's top industrialists (Cheema, Munir, and Su, 2016; Saeed, Belghitar, and Clark, 2015).
In these elections, a large number of Parliament members belonged to families that owned large business firms and groups, such as the Ittefaq Group of Nawaz Sharif family, current prime minister of Pakistan while chief minister of Puajab too belongs to this family. Besides, there are business groups such as Saif Group of current Senator Usman Saifullah, Salim Saifullah Kllan, and Anwar Saifullah Khan, former Provincial and Federal Ministers; JWD group of Jahangir Khan Tareen, former Federal Minister and current member of National Assembly; Service Industries of Ahmad Mukhtar,
fo1mer Defense Minister, Mirza Sugar Mills of Zulfiqar Mirza, former Interior Minister of Sindh Province; Masood Textile Ltd of Shahid Nazir, member of the National Assembly during 2002-2007 and Tandlianwala Sugar Mills Ltd of Haroon Akhtar Khan and Humayun AkJ1tar Khan, former Ministers, and Media times and Worldcall of Salman Taseer, former Governor of Punjab (Saeed, Belghitar, and Clark, 2015). In a country where businessmen are significantly involved in politics and politicians have their own business interests, it is difficult to separate business interests from the interests of the state. Traditionally, it is understood that politically influenced firms misuse resources on a large scale through the abuse of their political influences and state power.
Reviewing the previous literature notably, there is a clear evidence which indicates the relationship between political influences and earnings quality (e.g., Sadiq and Othman, 2017; Chi, Liao, and Chen, 20 l 6; Al-dhamari, and Ismail, 2015; Liu, Saidi, and Bazaz, 2014; Chaney, Faccio, and Parsley, 2011 ). Some authors argue that politically influenced firms are more involved in manipulating earnings, and reporting poor earnings quality. For example, Sadiq and Othman (2017); Al-dhamari and Ismail (2015) find that politically influenced firms have lower accruals quality; they are more engaged in earnings smoothing (Liu, Saidi, and Bazaz, 2014), and reporting less transparent earnings (Riahi-Balkaoui, 2004). Other researchers opine that politically influenced firms are involved in income-decreasing earnings management activities in order to report less taxable income (Li, Selover, and Stein, 2011; Ramanna and Roychowdhury, 2010), and hiding political gains which they obtain through dubious legality (Sadiq and
8
Othman, 2017; Chaney, Faccio, and Parsley, 2011 ). However, to the best of knowledge, none of the study has investigated political influences and earnings quality using three aspects of political influences and three attributes of earnings quality, which includes both accruals and real earnings management attributes of earnings quality. This is because; including only one attribute i.e. accruals or real earnings management underestimates earnings management activities of a firm. Therefore, this study adopted both accruals earnings management and real earnings management attributes in order to estimate and capture all earnings management activities of a firm, and thus gives more insight on the relationship between political influences and earnings quality.
Pakistan has numerous unique characteristics of politically influenced finns, which provided the opportunity to examine the impact of political influences on earnings quality in Pakistan. First, prior studies have used political influenced firm in a limited sense, where firm is assumed to be a politically influenced if it is affiliated with a single powerful politician (i.e. Dr. Mahathir and Dr. Anwar Ibrahim in Malaysia and Suharto in Indonesia). Unlike in Pakistan, there is a high presence of politicians and their close relatives in fmns. In fact, political system of Pakistan is controlled by influential families and business conglomerate (Saeed, Belghitar, and Clark, 2015), which is unlikely witnessed in previous studies. Secondly, congruent with the study of Cheema, Munir, and Su (2016), this study contributes to the agency theory and political economic theory by including another aspect of political influences (i.e. the presence of civil/military bureaucrats in a firm). This is because; Pakistan has directly been ruled by civil/military bureaucrats for couple of decades, which make these bureaucrats equally
influential as politicians. Thirdly, politically influenced firms of Pakistan provide a distinctive characteristic of data by offering almost equal nwnber of ruling and opposition party of politically connected firms. This is because, ruling party(s) has generally faced a strong opposition party(s) in Pakistan which provides the opportunity to investigate whether political connected ruling party firm report earnings quality poorer than a political connected opposition party firm. Fourthly, there is a strong nexus between politics and business in Pakistan, and numerous scandalous politically influenced firn1s are named of manipulating eamings as mentioned earlier. However, limited empirical evidence is available on the issue. Therefore, this study is motivated to give more insight on the relationship between political influences and earnings quality, and thus quantitatively investigated the impact of political influences on earnings quality in Pakistan.
1.2 Problem Statement
It has been observed that earnings quality is high when earnings numbers truly reflect firm's current performance (Dechow and Schrand, 2004); they are free from earnings management (Chaney, Faccio, and Parsley, 2011; Barth, Landsman and Lang, 2008), and contain high transparency (Sadiq and Othman, 2017; Bhattacharya, Daouk, and Welker, 2003). Consequently, financial statement users can make a well informed decision when firms are reporting high earnings quality, which reflects true and fair value of the firm (Dechow and Schrand, 2004). The above scenario is not met when firms' managers are engaged in earnings management to achieve desirable results but end up with lower earnings quality.
10
Managers often involve in earnings management to hide and cover the real performance of a firm (Braam, Nandy, Weitzel, and Lodh, 2015; Chaney, Faccio, and Parsley, 2011;
Teoh, Welch, and Wong, 1998). Consequently, the occurrence of earnings management deteriorates the reliability of earnings numbers, which results in unpredictable, non- sustainable, and less transparent earnings.
The above scenario gets worse when firm is politically influenced, creating agency problem and revealing that politically influenced finns are more involved in earnings management activities and reporting poor earnings quality (Sadiq and Othman, 2017;
Chi, Liao, and Chen, 2016; Braam, Nandy, Weitzel, and Lodh, 2015; Chaney, Faccio, and Parsley, 2011 ). This is because politically influenced finns are involved in tax- induced earnings management activities (Li, Selover, and Stein, 2011 ), obtaining political benefits through dubious legality (Chaney, Faccio, and Parsley, 2011 ), and then cover those political benefits by getting involved in income-decreasing earnings management activities (Ramanna and Roychowdhury, 20 I 0), and thus report poor earnings quality.
Pakistan is not an exception in respect of political influences and earnings quality; there is a nexus reported between politics and business; there are scandalous firms and companies as mentioned above that direct and monitor most of the political and business fortunes of the country.
Due to seriousness of the issue and consequent to its impact on shareholders, researchers have suggested studies to be conducted to examine the impact of political influences on earnings quality. For instance, Sadiq and Othman, 2017; Chi, Liao, and Chen, 2016; Al-dhamari and Ismail (2015) suggest that political influences and earnings quality need to be investigated in more detail using different aspects of both variables. Although there are studies that have investigated the correlation between politically influenced firms and earnings quality, very little focus was given to investigate the relationship between politically influenced firms and earnings quality in comparison to non-politically influenced finns using multiple aspects of political influences and multiple attributes of earnings quality.
Given the importance of the issue, seriousness of the consequences and impact of the issue to many stakeholders, this study will use three aspects or types of political influences simultaneously to investigate political influences and earnings quality: first, presence of politician(s) or close relative(s) of a politiciao(s) in board of directors or senior management; second, significant government shareholding in a firm; third, presence of fo1mer/current civil/military bureaucrat(s) in board of directors or senior management) and three attributes of earnings quality (i.e. Real earnings management, accruals earnings management, and earnings aggressiveness). The regulatory bodies and conflict of interest laws are weak in Pakistan, which allow powerful politicians and their close relatives, and civil/military bureaucrats to manipulate things and obtain private gains (Cheema, Munir, and Su, 2016).
12
As mentioned, the impact and consequences of the issue is tremendous. Political influences are considered to be an important detenninant of the Asian economies (Harymawan and Nowland, 2016; Chi, Liao, and Chen, 2016; Gul, 2006; Fisman, 2001), and particularly in Pakistan (Sadiq and Othman, 2017; Cheema, Munir, and Su, 2016; Saeed, Belghitar, and Clark, 2015; Khwaja and Mian, 2005). Considering the issue, current study provides a detailed investigation regarding the relationship between political influences and earnings quality in Pakistan, because Pakistani finns provide diversity in the data of politically influenced firms. In Pakistan, there is a high presence of civil/military bureaucrats, politicians from ruling/opposition parties and their closed relatives in finns, which offers useful variation in terms of the type of political influences that is used to examine the phenomenon. The study used politically influenced and similar non-politically influenced listed firms in Pakistan Stock Exchange (PSE) in order to get a clear picture of earnings quality in each of the finns, and to quantitatively investigate in detail the effects of political influences on earnings quality.
1 .3 Research Questions
Based on the above research problem, which is to provide a detailed investigation regarding the relationship between political influences and earnings quality in Pakistan.
This study addressed a few fundamental questions:
1) Is there a negative relationship between political influences and earnings quality?
2) Does a politically connected ruling party firm report earnings quality poorer than a politically connected opposition party firm?
3) Do politically influenced finns substitute between earnings management strategies?
1.4 Research Objectives
This study used politically influenced listed companies of Pakistan in order to get a clear picture of earnings quality in each of the firm, and to quantitatively investigate the effects of political influences on earnings quality. To achieve this, the study has general and specific objectives. Following are the objectives of the research:
1) To examine whether there is a negative relationship between political influences and earnings quality; to judge whether politically influenced firms report poorer earnings quality as compared to non-politically influenced firms.
2) To examine whether politically connected ruling party firms are reporting poor earnings quality compared to politically connected opposition party firms.
3) To investigate whether politically influenced finns substitute real earnings management for accruals earnings management.
14
1.5 Significance
The impact of political influence on earnings quality remains an area of interest to policy makers, accounting practitioners, auditors, and academic researchers. Though, there are a number of studies that have investigated the impact of political influence on earnings quality, recent studies such as Chi, Liao, and Chen (2016); Al-dhamari and Ismail (2015); Zeng, Li, and Hong (2012) have recommended that political influences and earnings quality need to be investigated in more detail and critically. Hence, this study provides a useful contribution to the existing literature as it has examined the relationship between political influence and earnings quality using multiple aspects and attributes of both variables.
First, this study contributes to agency theory by using political aspects of agency theory.
Agency theory explains the basis of conflict that can occur between the agent (managers) and principal (shareholders). The conflict between the agents and principals occur when the act of agents do not meet the interests of the principals and thus create agency conflict. Previous studies show that the presence of politically influenced people in a finn creates agency conflict. However, prior studies argue on agency theory by using only one aspect of political influences i.e. the presence of politicians or close relatives of politicians in a firm. While current study contributes to agency theory by adding two more aspects of political influences (i.e. the presence of civil/military bureaucrats in a finn and a finn significantly owned by government). In addition, our findings contribute to the political economy theory, which argue that a powerful individual (through any aspect of political influences) maintain their self interest
(through targeted information disclosures) to the detriment of those individuals without power (Deegan, 2006).
Secondly, most of the previous studies have used data from multiple countries, while current study used only Pakistani firms' data, which overcomes some criticisms of multiple-countries data. Pakistan is famous for its doubtful relationship between politics and businesses (Saeed, Belghitar, and Clark, 2015; Khwaja and Mian, 2005). Over the last 25 years, three elected governments were dissolved on allegations of corruption and political abuse of power (Cheema, Munir, and Su, 2016). Due to the fact of the weak government institutions and regulatory bodies, the Pakistani business environment looks particularly weak to political influences (Khwaja and Mian, 2005). Pakistan has many politically influenced listed firms; however, there are no restrictions to such political influences (Saeed, Belghitar, and Clark, 2015). Such features of the Pakistani economy provided an interesting platform to examine the effect of political influences on finns' earnings quality. Therefore, this study suggests policy makers and regulators of Pakistan m particular and of other developing economies m general with similar socioeconomic conditions and comparable regulatory mechanism, in order to improve the quality of earnings in politically influenced firms.
Further, this study contributes and extends previous literature in several ways, by selecting Pakistani firms' data. Pakistan lies in the developing economies, which provides diversity in the politically influenced firm's data. Prior studies have used political influenced finn in a limited sense, where firm is assumed to be a politically
16
influenced if it is affiliated with a single powerful politician (i.e. Dr. Mahathir and Dr.
Anwar Ibrahim in Malaysia and Suharto in Indonesia). Unlike in Pakistan, there is a high presence of politicians and their closed relatives in firms. In fact, political system of Pakistan is controlled by influential families and business conglomerate (Cheema, Munir, and Su, 2016; Saeed, Belghitar, and Clark, 2015). Moreover, a political influence varies with the political strength, which provides useful variation in terms of the type of political influences that is used to examine the phenomenon. In addition, this study contributes to the existing literah.lre of political influences by including civil/military bureaucrats as another aspect of political influences. This is because;
Pakistan has directly been ruled by military and bureaucrats for several years, which make these bureaucrats equally influential as politicians (Cheema, Munir, and Su, 2016). In addition to civil/military bureaucrats, current study enrich the existing literature of political influences by including significant government owned firm as another aspect of politically influenced firm, because government will place their representatives on board and management to protect government's political interest and concurrently government also act as a shareholder.
Thirdly, this study examined the relationship between political influences and earnings quality using three aspects of political influences and three attributes of earnings quality. In addition, this study also investigated the difference in effect between ruling/opposition party political connections and direct/indirect political connections on earnings quality. Further, it empirically examines whether politically influenced firms substitute between earnings management strategies, which shall also be a
methodological contribution to this domain. Therefore, this study provides more insight and detail investigation on the relationship between political influences and earnings quality, by testing almost all possible aspects of political influences (in Pakistani context) and earnings quality (using both accruals and real earnings management attributes of earnings quality).
Fourth, the current study expands the literature by showing that politically connected firms related to opposition party are reporting poor quality of earnings compared to political connections related to ruling party, because connected finns related to opposition party are more involved in real earnings management. Notably, these findings are one of the main contributions of the current study to the body of literature of political influences and earnings quality. While previous studies have suggested that the politically connected firms related to ruling party are more involved in accruals earnings management (Chaney, Faccio, and Parsley, 20 l 1 ). The current study findings are contrary to the findings of Chaney, Faccio, and Parsley (2011). This is because;
Chaney, Faccio, and Parsley (2011) only employed accruals attributes of earnings quality in their study. However, this study has employed both accruals and real earnings management attributes of earnings quality.
After the implementation of Sarbanes•Oxley Act (SOX), firms shift their manipulation activities from accruals earnings management strategy to real earnings management strategy (Kothari, Mizik, and Roychowdhury, 2012; Cohen and Zarowin, 20 IO; Cohen, Dey, and Lys, 2008), particularly in politically influenced firms (Chi, Liao, and Chen,
18
2016; Braam, Nandy, Weitzel, and Lodh, 2015), because real earnings management activities are not easily detectable (Zang, 2011; Roychowdhury, 2006; Gunny, 2005).
These shifts of earnings management strategies were noticed in developed economies, where institutions and regulatory bodies are strong. However, the current study expands literature by showing that politically influenced firms manipulate earnings through both accruals and real earnings management measures in a less developed country (i.e.
Pakistan), where regulatory bodies are weak.
Fifth, this study extends literature by showing that the presence of politically influenced people in a firm creates agency problem, which resulted in reporting lower earnings quality. Hence, to reduce the impact of agency problem, strong regulatory bodies and enlightened shareholders are needed. The results of the cun-ent study will enable shareholders to evaluate the negative effects of political influences on reporting earnings quality. Further, other stakeholders can also use the findings of the current study to identify which factors should be taken into consideration when evaluating a finns' earnings quality, and thus corporate political influences should be taken into consideration in the valuation process. The findings of the current study contributed to in setting up the foundation of agency theory that explains the relationship between political influences and earnings quality in Pakistan and worldwide. Thus, the agency cost in political influences and earnings quality literature could extend to other developing and under-developing economies in the future.
1.6 Scope of the Study
The scope of this study is politically influenced listed firms in Pakistan. There are many reasons for selecting Pakistan data in this research. First, Pakistan's business and economy is very much influenced by politicians (Cheema, Munir, and Su, 2016;
Khwaja and Mian, 2005). Politically influenced firms offer a suitable condition to examine the quality of earnings in this environment. Second, the political and business culture in Paki~tan is unique in nature. Finally, focusing on one country will overcome some of the constraints of cross-country studies, such as possible difficulties in data collection, analysis and interpretation that arise as a result of differences in-cultural and legal system and like.
The study covers the seven-year period from 2008-2014. This period includes one lagged year and one lead year data, because the accruals quality models and real earnings management models adopted in the current study required one lagged year and one lead year data. Data were collected from the State Bank of Pakistan (SBP) and annual reports of 129 listed firms. Annual reports used for the study were retrieved from the website of firms. The analyses of this study focus on political influences and earnings quality, whicb cover one tenure of the elected government between 2009 and 2013.
I. 7 Organization of the Study
The content of this research is divided into six chapters. First chapter highlights the various important contents that are relevant to this study. Specifically, this chapter sheds
20
a light on the background of the study, which includes discussing earnings quality and its related scandals in the world and particularly in Pakistan. It also includes political influences and its impact on earnings quality. This chapter also explains the problem statement; it reports research questions and research objectives of the study; justifies significance of the study and discusses scope of the study.
The next chapter investigates the results of previous studies that are related to earnings quality and political influences. It consists of two main sections namely; earnings quality definitions and its measurements and political influences, and its impact on earnings quality. Chapter three explains methodology of the study. In addition, it also discusses underpinning theory i.e. agency theory and a theoretical framework; it discusses the hypotheses developed for this study followed by justification of variables, models specification, sampling and methods of data collection and finally methods of analysis.
Chapter Four discusses descriptive statistics, diagnostic tests, reporting of main results followed by sensitivity analysis. Chapter Five covers discussion of the results. The last chapter; Chapter Six, concludes the study by presenting contributions of the study, limitations of the study followed by future research, suggestion and recommendations.
CHAPTER TWO LITERATURE REVIEW
2.1 Earnings Quality Definitions
Earnings quality is contextual and it means different things to different earnings information users (Dechow and Schrand, 2004 ). For example, standard setters or regulators view earnings to be of high quality when it is in accordance with Accounting Standards. In contrast, when earnings are easily convertible into cash then it is viewed as high quality from creditors' perspective. Researchers have considered earnings quality in different ways (Kamarudin and Ismail, 2014 ). For example, Abdelghany (2005); Schipper and Vincent (2003) define earnings quality from the decision usefulness concept. The decision usefulness is the only vital attribute of high earnings quality, because it is empirically good and captures the intent of standard setters (Schipper and Vincent, 2003).
Dechow and Schrand (2004) describe earnings quality wider than decision useful concept, having three categories: first high earnings quality reflect finns' current performance; second, high earnings quality should be a good indicator of future operating perfonnance; and last, it is a useful measure for assessing firm value. In short, high earnings quality accurately annuitizes the real value of the firm. Similarly, Dechow, Ge, and Schrand (2010) in their study define earnings to be of higher quality when it gives more information about future operating perfonnance of a firm. Similarly, some researchers found persistence earnings as important measure of earnings quality.
22
For instance, Comiskey and Mulford (2000) define high earnings quality as the ability to generate sustainable earnings while Penman and Zhang (2002) classify sustainable future earnings as high quality. It is argued that earnings which are closer to cash flow and contain comparatively small amount of accruals, are considered to be of higher quality (Visvanathan, 2006).
It is argued that earnings which timely recognize losses are considered to be of higher quality of earnings (Basu, 1997). Some researchers such as Ashbaugh, Collins, and Lafond (2006) explain high earnings quality when earnings include timeliness and value relevance. Earnings quality problem arises when film's earnings lack transparency and include unusual items (Bhattacharya, Daouk, and Welker, 2003).
In addition, Barth, Landsman and Lang (2008) in their study state that earnings which involve less earnings management, timely recognizes losses and thus earnings is considered to be of high value relevance. Similarly, Guay, Kothari, and Watts (1996) argue that managerial opportunism decreases earnings quality and information accuracy. In contrast, Dechow and Schrand (2004) state that some finns will have low earnings quality because of the nature of their business, even without earnings manipulation. This will include firms working in unstable environments as well as high- growth firms. Consequently, both types of firms due to their nature of business and opportunistic management can find their earnings quality affected. In this regard, some authors opine that earnings information consists of errors made by the accounting system and fundamental earnings process (e.g., Dechow, Ge, and Schrand, 2010).
Similarly, Francis and Ke (2006) reveal that earnings quality issue arises from both business environment, and financial reporting processes, such as governance activities, reporting standards and managerial judgment.
In addition, the conceptual Framework of Financial Accounting Standard Board (FASB) gives a general guideline for high earnings quality (Schipper and Vincent, 2003). The main purpose of published accounts or financial reporting is decision usefulness, and thus financial statements which are useful for decision making are considered as high quality (Dechow and Schrand, 2004). Two characteristics behind the financial reporting purpose are representational faithfulness and relevance. Schipper and Vincent (2003) stated that the information is faithfully represented if it is free from error, complete and relevant, and it is said to be relevance if it has predictive or confirmatory value.
2.2 Measurements of Earnings Quality
As discuss above, earnings quality is a multifaceted concept and its measurement is also multifaceted. The current study measures earnings quality through multiple measurements such as accruals earnings management, real earnings management, earnings smoothing, and earnings aggressiveness. Generally, high earnings quality has a high level of persistency, less volatility and total accruals, have lower level of accruals and real earnings management. The following subsections discuss each of the measurement in detail.
24
2.2.1 Earnings Management
Earnings management is the process of getting desirable earnings through manipulations, within the limitations of generally accepted accounting practice (Davidson, Stickney and Weil, 1985). Similarly, Healy and Wahlen (1999) define earnings management as "intervention in the financial reporting process, with the intention to get some personal or private gain". Managers have the discretions to transfer current income to future periods without the violation of general accepted accounting principles (Dechow and Schrand, 2004), and accounting standards give discretions to managers in preparing financial reports (Kamarudin and Ismail, 2014).
Prior research provides evidence that earnings management exists due to the main functions and roles played by financial reporting. For instance, Bauwhede (2001) claims that some managers' get involved in earnings management activities due to their need for external financing, board quality, audit quality and taxes. Healy and Wahlen (1999) state that firms manage earnings to affect stock prices. Prior to seasoned equity offerings (SEO) firms manage earnings to increase share prices (Dechow and Schrand, 2004). The study also reveals that earnings management mainly occurs after an increasing performance and period of high economic growth in booming economies.
Whereas, Burgstahler and Dichev ( 1997) state that management manage earnings to meet benchmark. Similarly, Degeorge, Patel, and Zeckhauser (1999) reveal that at some places a few finns involve in earnings management to improve their financial image by trying to meet the expectation of investors and achieve profit forecast of the analyst.
High earnings quality exists in the absence of earnings management since unmanaged earnings are of higher quality compared to managed earnings (Dechow and Schrand, 2004). Levitt (1998) argues that earnings quality decreases when firm's management is involved more in earnings management.
Melumad and Nissim (2008) also state that earnings management does not necessarily mean fraud. Fraud though is a severe case of earnings management, whereas Levitt (1998) states that earnings management is found to be in the middle of legitimate accounting and absolute fraud. Badertscher (2011) on the other hand, divides earnings management in three categories: first, earnings management that violates accounting principles; second earnings management in the middle of legitimate accounting and violating GAAP, and third, earnings management within the requirement ofGAAP.
Badertscher (2011) too states in his study that firms have two ways to manage earnings.
First, firms manage earnings through accruals to get the desired level of earnings. This method is known as accruals earnings management (AEM). Second, firms manage earnings through changes in real business transactions, such as decreasing prices to increase sales, reducing discretionary expenses to increase current earnings, and overproducing goods (inventory) to reduce costs of goods sold. Managing earnings through real business transaction is known as real earnings management (Roychowdhury, 2006).
26
Zang (2011) opines that firms choose earnings management strategy based on its relative cost. It means that real earnings management (REM) is more useful compared to accruals earnings management (AEM) if the cost of using AEM relative to REM is higher and vice versa. The cost of manipulating earnings depends on the limitations that are applicable on using AEM and REM.
Cohen, Dey, and Lys (2008) also show that after the implementation of Sarbanes Oxley Act (SOX), the level of real earnings manipulation has increased, while accruals based earnings manipulation has decreased. In contrast, throughout seasoned equity offering (SEO) firms use both earnings management strategies (Cohen and Zarowin, 2010). The study further reveals that the SEO firms' inclination to use real earnings management activities is positively related to the costs of accruals earnings management in these firms.
Furthermore, Fields (2001) states that examining only one earnings management proxy at a time underestimates earnings management activities in a firm. Particularly, if managers employ both earnings management strategies as substitutes for each other, then examining onJy one strategy in isolation underestimates earnings management activities. Similarly, Braam, Nandy, Weitzel, and Lodh (2015) state that taking only accrual-based measure underestimates the earnings management activities. Therefore, this research is motivated to use both accruals and real earnings management as a measure of earnings quality.
2.2.1.1 Real Earnings Management
Firms manage their earnings to obtain the desired level of earnings, and managing earnings through real business activities is known as real earnings management (Roychowdhury, 2006). The Roychowdhury defines real earnings management as
"deviation from normal business practices, which is motivated by managers' desire to mislead some stakeholders into believing that certain financial objectives have been met in the normal course of operations". These deviations enable managers to meet financial reporting goals, even though they do not contribute to fim1 's economic value. Certain real earnings management practices, such as price discounts, overproduction of goods and reduction of discretionary expenditures are probably favorable measures in certain economic situations. However, managers are involved in real earnings management, if they are engaged more extensively in these activities with the objective of beating/meeting an earnings target, (Roychowdhury, 2006).
Graham, Harvey, and Rajgopal (2005) find in their sw-vey that managers are willing to manage real earnings to meet earnings targets such as analyst forecasts, zero earnings and previous periods' earnings, although real earnings management potentially reduces firm value. Real earnings management adversely affects finns' value because real operating measures taken in the current period to increase earnings can have an adverse effect on future periods' cash flows. For instance, to meet short-term earnings target finns offer price discounts to increase sales volumes which can lead customers to expect such discounts in future periods as well. In other words, it means lower profit margins on future sales.
28
Despite the costs linked to real earnings management (REM) and also because it imposes greater long-term costs on the finn value, managers are unlikely to rely only on accrual-based earnings management to manage earnings. In respect of loss linked to REM, Graham, Harvey, and Rajgopal (2005) state that fifty percent of the CFOs in their survey say that they would accept small losses to meet current earnings targets by postponing new projects, while eighty percent of the CFOs say that, they would decrease expenditure on advertising, maintenance and R&D to meet target earnings.
Roychowdhury (2006) on the other hand, give evidence that managers avoid reporting annual losses by decreasing selling, general and administration expenses, overproducing inventory goods to decrease cost of goods sold (COGS), and decreasing research and development
